I can't be sure from the photo, but his PS appears to have an
ATX connector. I may be wrong.
Addendum: Since the OP mentioned having an P8 connector, I doubt the
picture he linked to is for his PSU. That auction is for an ATX-style
PSU but no P8 connector is shown. Since the PSU is not inside a case,
I would think George could simply look at the PSU to see what was its
brand and model. There should be some labels on the PSU.
He did mention a green-colored wire. The old AT-style P8 and P9
connectors didn't have a green wire.
First, that's not a protection mechanism but just a different
method of turning on the power. AT was an IBM PC standard,
and IBM simply specified a hard switch rather than logical,
or soft, switch for the power, unlike the ATX standard.
Second, whether or not a supply needs a load to run has
nothing to do with its level of quality but with its design. Few
XT and AT supplies I tested would run without a load on the
+5V, including high quality IBM, Astec, and Delta. One of
the few that would run without load had the smallest transformer
and heatsinks of any in the power range and no UL approval..
The cable harness from the old AT-style PSU had 4-pin Molex connectors
(for 5V and 12V power to drives and such), the 4-pin mini connectors
(mostly for floppy drives and some tape drives), the hard-wired
switch, and the P8 and P9 motherboard connectors. From the pinout
shown at
http://pinouts.ru/Power/MotherboardPower_pinout.shtml, which
wire do you see that would have the mobo tell the PSU to power up?
Might it be the orange "voltages stabilized" wire (aka the "power
good" line)? It's been around 7 years since I worked on an box with
an AT-style PSU and mobo. A "stabilized" signal could only provide a
"good" signal *after* all the voltages came up, so the PSU would have
to come up first and then the mobo maybe decided whether it stayed up
or not. Rather than have the mobo tell the PSU that it could come up,
the mobo could tell it to power down; otherwise, how could the orange
wire for stabilized voltages work if the voltages weren't yet at the
mobo? You can't tell anyone the power is good until you get the
power. When I looked at
http://www.technick.net/public/code/cp_dpage.php?aiocp_dp=cir_ps_at_001,
the P.G. (Pwr_Good) line is an output, not an input, to tell the mobo
when the voltages are okay. Since the orange Pwr_Good wire is an
output, just like the other pins for P8 and P9, from the PSU, just
where does a "soft" control line go back to the PSU? In this case,
Pwr_Good seems to be nothing more than a decayed signal off the +5V
tap. See
http://www.pcguide.com/ref/power/sup/funcPowerGood-c.html.
According to
http://computer.howstuffworks.com/f...http://www.pcguide.com/ref/power/sup/func.htm,
the AT mobo didn't have a "soft" power control signal back to the PSU.
It used the hard-wired switch. I don't have an old AT laying around
to look at but this fits my recollection. I even remember having to
cut out and splice in new Power button switches on a couple of old ATs
to get them to power up since if they had a "soft" power up function
then the hard-wired switch was superfluous and you could bring up the
computer without it. There is no "soft" power on control from the
mobo back to the PSU in an AT-style system.
After a bit of searching, apparently some ATX-style PSUs include a P8
connector. Never seen one but do see them described by a few users'
posts. Don't know what would be the point unless they were letting
users connect an ATX-style PSU to an old AT-style motherboard (but
then what about the missing P9 connector?). I suppose before ATX took
off that there were these mutant PSUs that tried to bridge between the
old AT mobos and the new ATX ones.
Back to the OP, and since he probably does have an ATX-style PSU, and
since he shorted PS-On (pin 14) to a ground wire (black) and the fan
didn't come on, my guess is that he didn't put a load on the PSU, like
hooking up a hard drive or power resistor (8-ohm, 20-watt should work;
http://snipurl.com/sq6t) across the +12V. The PSU won't come on or
will shut down with too little load (because the PWM controller can't
provide the small duty cycles or pulses to the switchmode transistor
needed for low loading.).
If you look at the schematic of an example ATX-style PSU at
http://www.pavouk.comp.cz/hw/en_atxps.html, you can see why the PS-On
line is pulled up to +5V (when the 20-pin connector isn't on the
mobo). The PS-On line is pulled up to the +5VSB input through
resistor R23. If and only if the 20-pin connector is attached to the
mobo can the PS-On signal get pulled down by logic on the mobo when
the user presses the Pwr button.
The feedback line provides overvoltage protection, so I suppose the
voltage (at R25 and R26 between the 12V and 5V) is out of range when
the taps are unloaded. (Note: I just happened to find this schematic
and cannot attest to the quality of the unit.) According to
http://www.kellerstudio.de/repairfaq/sam/smpsfaq.htm, "In most
designs, the +12 and -12 V supplies merely track the 5V supply, and
are not separately regulated. They may soar to higher voltages anyway
if unloaded, but will be additionally increased in voltage by the
ratio of 5V output increase. Even though the rating of the 5V
electrolytic may not be exceeded, and still have a sufficient safety
margin, this may not be the case for the 12 V outputs." Also noted at
http://repairfaq.cis.upenn.edu/Misc/smpsfaq.htm was, "Note that an
underloaded supply may be cycling due to overvoltage and there may
actually be nothing wrong! Many SMPSs (switchmode power supplies)
require a minimum load to maintain stability and to provide proper
regulation. This is typically 20 percent of maximum on the primary
output (the one which drives the feedback loop). However, minimum
loads may also be needed on other outputs depending on design. The
only way to be sure is to check the manufacturer's specs." While you
and I know the overvoltage is because the outputs are unloaded, the
PSU won't know that so it is trying to protect any devices that might
be connected to it. The feedback circuit may fail. There might not
even be one (and why it is a "cheapy" unit not based on price but poor
design). Without a load, the ATX-style PSU should not turn on. The
OP never mentioned putting a load on the PSU.
